See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/233996026

Scaffold Development and Characterization Using CAD System

Article · October 2011

DOI: 10.5099/aj110400268

CITATIONS READS
6 351

4 authors, including:

Ahmed Farag Seddik Ahmed Agameia

Thebes Academy Helwan University
77 PUBLICATIONS   325 CITATIONS    4 PUBLICATIONS   9 CITATIONS   

SEE PROFILE SEE PROFILE

M A Sharaf
Helwan University
33 PUBLICATIONS   390 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Design and Development of Optical Device for Blind People View project

Dielectrophoresis View project

All content following this page was uploaded by Ahmed Farag Seddik on 31 May 2014.

The user has requested enhancement of the downloaded file.

 American Journal of
Biomedical Sciences
ISSN: 1937-9080
nwpii.com/ajbms

Scaffold Development and Characterization Using CAD System

Ahmed Farag Aly1*, Ahmed Agameia1, Amal Samir Eldesouky1, and Mohamed A. Sharaf2
1
Department of Biomedical Engineering, Helwan University, EGYPT.
2
Department of Chemistry, Helwan University, EGYPT.
*Corresponding Author:
Dr. Ahmed Farag Aly
Associate Professor
Department of Biomedical Engineering
Helwan University
EGYPT
Email: afarag@mcit.gov.eg

Received: 19 October 2010; | Revised: 9 March 2011; | Accepted: 11 August 2011

Abstract
Morphology and mechanical properties of scaffolds seeded with osteoblastes cells used for bone and
cartilage repair are the critical factors in bone tissue engineering. In this work, adding CMC and controlling
temperature for nano-hydroxyapatite (HA)-b-tricalcium phosphate (b-TCP) scaffold using Polymeric sponge
method provide suitable properties. A developed computer system was used to determine properties of
scaffold. Porosity, shape and connectivity of pores were analysed based on image processing method. Cells
were seeded on scaffold and the differentiation rate was calculated using image analysis. The fabricated
sample showed high porosity (nearly 61%) and high compressive strength (nearly 16 MPa), as well as having
a well pore size of 200 μm and more. Comparing to Archimedes method, the image result was more
accurate. Internal porosity was more than surface porosity due to skin effect.

Keywords: Biomaterial, computer aided system, interconnection, porosity, morphology.

bone tissue engineering must be biocompatible,

1. Introduction
Composites biodegradable polymers and
bioactive ceramics in the form of biocompatible
scaffolds have been proposed and used in the field
of tissue engineering with emphasis towards hard
tissue regeneration [1]. The generation
biomaterials are expected to be degradable and
bioactive porous scaffolds for the living bone
tissue engineering reconstruction [1]. Scaffold for
osteoconductive, and have a macroporous
structure [2]. Hydroxyapatite (HA) bioceramic
have been found to bond with living bone tissue
through a bone-like apatite layer due to its good
biocompatibility, osteoconductivity and a similar
mineral composition to natural bone [3]. Due to its
brittleness, it isn’t recommended for load bearing
scaffold. B-TCP is chemically stable and has a
fairly fast bioresorption rate [4]. TCP is regarded
as an ideal bone substitute. carboxymethyl

Am. J. Biomed. Sci. 2011, 3(4), 268-277; doi: 10.5099/aj110400268 © 2011 by NWPII. All rights reserved. 268
cellulose (CMC) can react strongly with CS and
act as an ionic cross-linking agent at the
appropriate pH [5].
In order to understand the behaviour of
scaffold, quantitative knowledge of the
morphology and the mechanical behaviour is
necessary. Recent advances in computing
technologies both in terms of hardware and
software have helped in the advancement of
design and analysis. Therefore, the porous
structures need to be quantified morphologically.
We prepared hydroxylapatite and tri-calcium
phosphate nano-particles. Different concentrations
of powder were used to get slurry. CMC was
added to provide strength. Sponge was immersed
in slurry and treated using temperature.
Mechanical properties, degradation behaviour and
cell proliferation were tested. Morphology was
detected using developed computer system.
Segmentation is a necessary step. In the past few
decades, many effective algorithms have been
proposed to perform the computer-aided
segmentation. Compared with the algorithms for
common image processing, this system is used for
porous bio-materials.
Also, images are usually influenced by noises
and partial volume effect [6], algorithms should be
sophisticated enough to handle the segmentation
task. Algorithms are classified into two categories:
algorithms based on threshold, algorithms based
on morphology operations.
2. Materials and Methods
2.1 Nano-particles Preparation
The basic materials, hydroxyapatite and tri-
calcium phosphate particles, were prepared in lab
using chemical precipitation method. The
precipitates are used to prepare scaffold. CMC is
used to strength scaffold and control rate of
degradation. Fig. 1 shows the procedure of Nano-
particles preparation.
2.1.1 Synthesis of Nano hydroxyapatite
Particles
Hydroxyapatite was prepared using chemical
precipitation method. 500 ml of 0.6M (NH4)2PO4
were added to DI water. 500 ml of 1 M calcium
nitrate solution was prepared by dissolving in DI
Am. J. Biomed. Sci. 2011, 3(4), 268-277; doi: 10.5099/aj110400268
water. The pH of the solution was adjusted to 11
by addition of NH4OH. White precipitates of
hydroxyapatite were formed by dropping
diammonium hydrogen phosphate solution to
calcium nitrate solution under constant stirring for
one hour. The solution was reflux for one hour and
aged for one day. The precipitates washed with DI
water, dried at 80oC for a day and calcinated at
800 oC for one hour.
2.1.2. Synthesis of Nano –tricalcium Phosphate
Tricalcium phosphate was prepared using
chemical precipitation method. Solution
containing 0.4M of (NH4)2PO4 was added to
calcium nitrate solution containing 0.6M of
Ca(NO3)2 _ 4H2O to make 1000ml of solution.
The pH of the solution was adjusted to 10 by
addition of NH4OH. White precipitates were
formed solution aged for one day. The precipitates
washed with DI water and ethanol dried at 80 °C
for another day then the temperature increased to
300 °C for three hours. The dried powder was
milled and calcinated at 900 °C for two hours.
dropping
Ca(NO3)2· 4H2O (NH4)2HPO4
sol.
sol.
Adjust PH by ammonium
sol.
Turbid sol.
Age one day
Washing
HA and TCP
Calcination particles
Figure 1. Preparation method of powders.
2.2. Polymeric Sponge Method
The polymeric sponge method is performed
by impregnating porous polymeric substrates
(sponges) with HA/TCP slurry. The slurry was
prepared by dispersing calcinated HA and TCP
powders in solution of poly vinyl alcohol (PVA)
(5 weight %) as the binder. Drops of CMC were
© 2011 by NWPII. All rights reserved. 269
added. The soaked sponge was oven dried at 400C
for 24 hours. The dried sponge was sintered at
650C for 3 hour at a heating rate of 5C per minute.
The samples were kept at 650C for one hour for
binder burn out. The sintered porous scaffold
characterized for strength, porosity, pore size
distribution and microstructure. The process flow
chart for fabrication is shown in Fig. 2.
2.3. Mechanical Test
The compressive strength was tested for the
three concentrations of HA and b-TCP with a
testmetric universal testing machine (AG-10AT,
Japan) with a compression strength at rate of 0.5
mm/ min_ at a room temperature of with a relative
humidity of 15 percent until 50% reduction in
specimen height.
HA/TCP
slurry Sponge
Dry at 85 for one day
Heat at 400 degree
for 2 h
Sinter at 650 for 3h
Figure 2. Polymeric sponge method.
The scaffold was cut into blocks with a size of
10 _ 12 mm. Three parallel samples were tested
for every scaffold and the mean value of the
compressive strength of different scaffolds was
given. The average data of the test results are
plotted. The elastic modulus was calculated as the
slope of the initial linear portion.
2.4. Scaffold and Image Analysis
Scaffold sample were imaged via scanned
electron microscope. The samples were mounted
on cupper stubs and coated with gold using coated
sputter (S150A Edward, England). The specimens
then were examined under JXA-840A electron
probe micro analyzer (Jeol,-Japan). Scanning
electron microscopy (SEM) was used to image
and determine the morphology of the scaffold
using our developed system (CAD system).
Am. J. Biomed. Sci. 2011, 3(4), 268-277; doi: 10.5099/aj110400268
Binary images of SEM images are obtained using
threshold. Segmentation is used for detecting solid
area of scaffold material and void area (pore).
Analysis of image is used to calculate total
porosity and interconnectivity. It also, provides
information for each pore and its pore size instead
of detecting by SEM. In CAD system, edge
detection for input image is available with
different techniques. Scaffold was trimmed to
approximately 10mm x10mm x15mm using
Buehler Isomet low speed saw. Thickness of
sample does not affect on porosity, since we
calculate surface porosity for each image and
foreground objects in it.
2.4.1. Image processing
Thresholding is the first step in image
analysis. Threshold-value is used to differentiate
between foreground and background.
Thresholding segments the input image into pores
(void area) and scaffold materials. Small pores
arising at the surface as a result of segmentation
are removed automatically by erosion without
affecting the bulk pore structure. Pore boundaries
were detected and objects that are close to each
other are connected. Subtracting from the
background gives edges and area of foreground
objects. By filling process, area of pores is easily
detected and eliminated small dark pixels in pores.
Each region in the resulting image is labelled to
extract the parameters for all pores in image. Fig.
3 illustrates the process for foreground detection.
Threshold for image
Input
image
Morphological operation
Hole filling & Object labelling
Object Properties
Figure 3. Summary for Image analysis technique.
2.4.2. Interconnectivity of Pores
In the calculation of the interconnectivity of
the pore spaces, the threshold was used. The area
of the interconnected spaces could be calculated.
© 2011 by NWPII. All rights reserved. 270
According to method of capture image,
interconnectivity is calculated.
For longitudinal images in scaffold, the area of
interconnected pores was divided by the total area
of the pore spaces, thus yielding the degree of
interconnectivity. The algorithm uses a
combination of morphological operations applied
to binaries images. By dilation operation, the
connected pore regions that are more than defined
area of pore are detected.
For transverse sections, interconnectivity is
calculated by subtracting two consecutive images
after detection of pores in each image. One minus
white area divided by area of pores in first image
represents the connectivity. Fig. 4 Shows steps of
determination interconnectivity in case of
transverse both and longitudinal sections.
1st layer 2nd layer
image image
Detection of pores in each image.
Subtracting two images
Division white area by area of pores in
first image. (A)
1 - A
Interconnectivity
Figure 4a. Interconnectivity using transverse sections.
2.4.3. Biodegradation Test
Biodegradation test of porous scaffold was
done using Archimedes principle. Biodegradation
test of porous scaffolds was done by soaking in
Tris-HCl buffer solution for one to four weeks
then the samples were dried. Weight of sample
after soaking in Tris-HCl solution was calculated
after each week. The result was compared to that
calculated by CAD system to verify the accuracy
of the developed program and show that it could
be used to calculate degradation rate.
From imaging of scaffold material before
treatment in simulated body fluid and after
treatment, degradation is calculated. Threshold is
Am. J. Biomed. Sci. 2011, 3(4), 268-277; doi: 10.5099/aj110400268
used to detect area of scaffold material in image.
The difference area of the two images is divided
by native area of scaffold material in first image.
This degradation could be measured for different
treatment time and curve indicated rate of
degradation is plotted.
Input
image
Detection of pores
Dilation
Calculation area of connected pores (A)
Calculation area of total pores (B)
A /B
Interconnectivity
Figure 4b. Interconnectivity using longitudinal section
2.5 Cell Isolation and Culture
Samples of blood were taken from cord
blood. 30 ml for sample aspirated from cord
blood.
The cord blood aspirated was diluted with
phosphate buffer saline (PBS). 15ml of Ficoll to
the aspirated sample was added and mononuclear
cells (MNCs) were separated. Cell centrifuged at
1500 rpm for 10 minutes to form a cell pellet. The
pellet was then resuspended in fresh medium
Alpha modified Eagle's medium (Alpha MEM).
The mononuclear cells plated in and cultured at
37°C in 5% CO2 incubator.
The scaffolds were seeded with cells by
dispensing the cell suspension directly on their
surfaces. The seeded scaffolds were incubated at
37°C and 5% CO2 in a humidified incubator.
Scaffolds were microscopically evaluated three
and eight days after seeding using SEM and Phase
contrast microscope.
© 2011 by NWPII. All rights reserved. 271
3. Material Characterization the area of empty space is obtained. The ratio of
white area to the total area of the image represents
3.1. Nano-particles Images the porosity [7, 8], i.e. dividing the area of the
The prepared particles were imaged by void empty space by the area of the ROI gave
transmission electron microscope. HA particles porosity. Determination of surface porosity using
were less than 50nm as shown in Fig.5a. For TCP, image analysis is more accurate than
the particles as shown in Fig.5b were in range of determination using Archimedes’ principle. The
200 to 300 nm. porosity obtained by image processing is around
the measured value by any other methods as
shown in Fig. 7. The comparison was done to
confirm the validity of the software. It shows that
the calculated porosity by CAD is greater than it
by Archimedes. This may be due to ability of
detecting small pores in case of using image.
Calculating porosity using Archimedes’
principle shows that it increases as powder
concentration decreases. Porosity of scaffold was
in range of 34 to 55 at surface.

Figure 5a: TEM images for HA

Figure 5b: TEM images for TCP.

3.2. Polymeric sponge method

Scanning electron microscopy shows the
morphology of the scaffold using sponge method.
Images show high porosity. Pore size’ range is
from 50 to 600 micrometer as shown in Fig. 6. A
wide range of pore diameters could be seen within
the same SEM image. Figure 6. Cross-sectional SEM micrograph of a
scaffold using sponge method with 15% powder
3.3. Porosity Calculation Based on Image concentration.
Analysis
Porosity is the simplest property It is found that porosity of scaffold doesn’t
characterizing a porous medium. Porosity is the change significantly with particle concentration
fraction of void space within the sample. After increases from 15 to 20%. But, it is more porosity
estimating the area of the scaffold within the ROI, with concentration 10 than it in 15%

Am. J. Biomed. Sci. 2011, 3(4), 268-277; doi: 10.5099/aj110400268 © 2011 by NWPII. All rights reserved. 272
concentration. Pore size is different as powder
concentration changed. It decreased nearly by 25
to 80 micrometer as powder concentration
increases.
Figure 7. Porosity using image and archimedes
principles.
Figure 8. Effect of threshold value in detection objects
3.3.1. Interconnectivity
Scaffolds show good connectivity. First all
pores are detected using threshold and dilation.
Sum of pores area is calculated. Pores with size
Am. J. Biomed. Sci. 2011, 3(4), 268-277; doi: 10.5099/aj110400268
more than 100 micrometers are calculated. The
calculated parameter (interconnectivity) was
ranged from 80% to 90%. The interconnectivity
not affected by changing powder concentration as
determined in other studied [9]. It is found that
threshold value is important for success of
analysis. Fig. 8 shows the effect of changing
threshold value from 0.3 to 0.5.
3.3.2. Morphological Process
Close operation detected edges of foreground
objects and connected objects that are close to
each other. Background subtraction with SE of 25
was shown in Fig. 9-a. Background subtraction
and filling holes result in detection of foreground
objects as shown in Fig. 9-b, it is seen that the
algorithms success in defining pores.
3.4. Mechanical Test
Samples of scaffold were tested and the mean
value of the compressive strength of scaffolds
with different powder ratio was measured. The
average data of the test results are plotted in the
Fig.10.
Compression strength increases with
decreasing the porosity and increasing powder
concentration. It was up to 20MPa. These values
were suitable and similar to those for bone. It is
found that as powder concentration increases,
compression strength increases. In comparison to
scaffolds using other methods, it is indicated that
sponge method using the controlled temperature
has higher compression strength. This is due to
polymer used and controlled temperature. The
roughness surface for scaffold was measured. It
showed high roughness values compared to other
materials. This indicated that cells will have high
adhesion degree to scaffold surfaces. Roughness
increased as nanoparticles concentration
increased. It was inversely with CMC
concentration.
3.5. Biodegradation Test
Samples showed a significant increase in the
mean weight loss percent throughout all week. All
three samples (s1, s2, s3) had high weight loss
percent during the first week (ranging from 35%
to 4o%). However, the weight loss slowed down
in the next weeks as shown in Fig. 11. Using
© 2011 by NWPII. All rights reserved. 273
images, high loss in scaffold material was found in
first week. This result indicates that sampled still
do its function and give time for cell to
differentiate up to 28 days.

Figure 11. Mean weight loss for three samples at the

end of each week.

3.6. Cell differentiation

The cells were used for scaffold seeding after
they differentiated in culture and reached near
confluence. Scaffold was imaged using scanning
electron microscopes. The images showed that the
osteoblastes cells adhered on the scaffolds’
surfaces. The images show good differentiation
and growth as shown in Fig.12. It shows cell
growth on scaffold surface after 7days. Cell
stretch itself across pore and cover the surface.

Figure 9. Subtraction background and Fill pores for

detection of foreground objects.

Figure 10. Compression and powder concentration.

Figure 12. SEM micrographs show cell growth on
scaffold surface after 7days.

Am. J. Biomed. Sci. 2011, 3(4), 268-277; doi: 10.5099/aj110400268 © 2011 by NWPII. All rights reserved. 274

Figure 13. SEM micrographs of the scaffold surface
shows cell growth after 14 days.
These results indicated that the implemented
scaffold is biocompatible. Scaffold allows cell
adhesion and differentiation.
Figure 14. SEM micrographs of scaffold-cell
constructs after seeding showing the attached cells
on the pore wall of the scaffolds [a] and cells filling
up the open pore by stretching themselves across the
pore.[b].
After two weeks of seeding, the cells were
differentiated inside scaffolds’ pores and move
out of pores. Cells continue differentiation as
shown in Fig.13. The image of the scaffold
surface shows cell growth with high dense after
Am. J. Biomed. Sci. 2011, 3(4), 268-277; doi: 10.5099/aj110400268
14 days. The cells penetrate into the scaffold and
migrate. The cells also attached to and colonized
the walls of the scaffolds’ pores as shown in
Fig. 14. The cells not only lined the walls, but
also stretched themselves across pores and
tunnels as shown in Fig. 14.b. Some areas of the
scaffolds’ surfaces became covered with a layer
of reticular deposits. After 28 days the scaffold
was nearly full degradable and structure like
bony architecture was formed as shown in
Fig.15.
Figure 15. SEM micrograph shows bony structure.
4. Summary
Three scaffolds were prepared containing the
same concentrations of PVA but different ratios of
nanoparticles. The ratio was 10-15 and 20%.
β­TCP to HA ratio was 2:3. Same CMC
concentrations were added to each group. The
scaffolds were imaged using scanning electron
microscope (SEM). Segmentation algorithm with
several techniques is used to determine
parameters. Morphology analysis has been
developed to determine the pore distribution and
porosity of porous structure. The system allows
user to define and change the used threshold
value. Evaluation of implemented scaffold using
CAD system is simple and fast. It doesn’t need
user experience as it draws boundaries.
The compressive strengths of the scaffolds
were measured by compressing samples to failure
using a mechanical testing machine. High
compressive strengths in scaffold with high
powder concentration are observed. In co-relation
with porosity, compressive strengths increase as
porosity decrease. It also shows that compressive
© 2011 by NWPII. All rights reserved. 275
strengths increase as CMC concentration
increases. The degradation rate of the scaffolds
was evaluated by comparing the scaffolds’ dry
weights before and after immersion in a simulated
body fluid. Image system is used to define
degradation of scaffold. Scaffolds were scanned
before and after treatment in simulated body fluid.
Image technique was used to compute difference
between the two images, so degradation rate is
computed. The results showed that all scaffold
groups experienced rapid weight loss during the
first week which slowed down thereafter.
Increasing in CMC and ceramic concentration
decreases the scaffold’s degradation rate.
5. Conclusions
Using a new treatment process for
implementing a scaffold by Polymeric Sponge
Method produces high porous scaffold with high
strength. These scaffolds are suitable for cartilage
and bone tissue engineering. Using developed
computer system permits scaffold characterization.
Based on the obtained results, the prepared
composite scaffolds overcame the disadvantage of
using HA or TCP only. Composite based on using
the two ceramic powders together allows
controlling of degradation and mechanical
strength. Adding CMC increased the mechanical
properties and resulted in a controllable
degradation rate. But, it decrease roughness value.
The scaffolds porosity ranged from 30 to 60%
which is similar to that in part of bone. The
scaffolds have a highly porous interconnected
microstructure. The increase in the powder
concentration in scaffolds decreases their pore size.
The increase in the powder concentration increases
the compressive strength of the scaffolds. The
degradation rate of the scaffold varies inversely
with the CMC and powder concentration. The
HA/β­GP/HEC scaffolds are biocompatible since
they allow the seeded cells and also encourage cell
attachment and differentiation.
The developed image analysis system (CAD)
is easy system and fast evaluation for any porous
material. It computes porosity, interconnection
directly from image. Measuring pore size is
allowed without need scan system of SEM
instrument. Using computer system characterises
Am. J. Biomed. Sci. 2011, 3(4), 268-277; doi: 10.5099/aj110400268
porous scaffold before and after seeding cells. It
could be used with other system as finite element
to provide complete evaluation, i.e. structure and
mechanical properties without need mechanical
tests. The SEM may yield digital images with
suitable quality for further analysis. The
acquisition of numerical information from these
high quality images requires good and effective
processing and analysis routines. Using successive
stages of erosion/dilation has revealed information
about the porosity, connectivity of pore structure
and allowed for the estimation of pore sizes.
References
1. D. Mohn, S.K.Misra, T.J. Brunner, A.R.
Boccaccini, W.J. Stark, Nano- versus Micron-
sized Bioactive Glass Reinforcement of
P(3HB) – Are Nano-fillers the Way Forward,
European Cells and Materials, Vol. 16., (page
8), ISSN 1473-2262, 2008.
2. Hirotaka Maedaa, Toshihiro Kasugaa,,
Masayuki Nogamia, Minoru Ueda, Preparation
of bonelike apatite composite for tissue
engineering scaffold, Science and Technology
of Advanced Materials 6 (2005) 48–53. DOI:
10.1016/j.stam.2004.07.003
3. Zhihong Dong, Yubao Li, Qin Zou,
Degradation and biocompatibility of porous
nano-hydroxyapatite/polyurethane composite
scaffold for bone tissue engineering, Applied
Surface Science 255 (2009) 6087–6091. DOI:
10.1016/j.apsusc.2009.01.083
4. Wang M, “Developing bioactive composite
materials for tissue replacement,”
Biomaterials, vol.24, pp. 2133-2151, 2003.
DOI: 10.1016/S0142-9612(03)00037-1
5. Liuyun Jiang, Yubao Li, Xuejiang Wang, Li
Zhang, Jiqiu Wen, Mei Gong,Preparation and
properties of nano-hydroxyapatite/chitosan/
carboxymethyl cellulose composite scaffold,
Carbohydrate Polymers 74 (2008) 680–684.
DOI: 10.1016/j.carbpol.2008.04.035
6. W. Sun, B. Starly, J. Nam, A. Darling, Bio-
CAD modeling and its applications in
computer-aided tissue engineering, Computer-
© 2011 by NWPII. All rights reserved. 276
 Aided Design 37 1097–1114, (2005). DOI:
10.1016/j.cad.2005.02.002
7. Lin C L, Miller J D., “Network analysis of
filter cake pore structure by high resolution X-
ray microtomography,” J. Chemical
Engineering Journal, vol. 77-7, pp. 79−86,
2000.
8. Coskun S B, Wardlaw N C., “Influences of
pore geometry, porosity and permeability on
initial water saturation: an empirical method
for estimating initial water saturation by image
analysis, ”J. Petroleum Science and
Am. J. Biomed. Sci. 2011, 3(4), 268-277; doi: 10.5099/aj110400268
View publication stats
Engineering, vol. 12-4, pp. 295−308, 1995.
DOI: 10.1016/0920-4105(94)00051-5
9. R.Ziel , A. Haus, A. Tulke, “Quantification of
the pore size distribution (porosity profiles) in
microfiltration membranes by SEM, TEM and
computer image analysis,” Journal of
membrane science, vol. 323, pp. 241-246,
2008. DOI: 10.1016/j.memsci.2008.05.057
© 2011 by NWPII. All rights reserved. 277